The theory of career mobility (Sicherman and Galor 1990) claims that wage penalties for overeducated workers are compensated by better promotion prospects. Sicherman (1991) was able to confirm this theory in an empirical study. However, the controls for the opposing phenomenon of undereducation used in his tests produced unconvincing results, for which no sound theoretical explanations were given. The only re-test yet conducted (Robst 1995) also produced ambiguous results. In the present paper, we estimate random effects models to analyze relative wage growth using data from the German Socio-Economic Panel. We find that overeducated workers have markedly lower relative wage growth rates than correctly allocated workers, while undereducated workers enjoy higher rates of relative wage growth. Our results cast serious doubt on the career mobility model, at least with respect to the overeducation issue. In view of the acknowledged positive correlation between access to training and upward career mobility, the plausibility of our results is supported by the finding that overeducated workers have less access to formal and informal on-the-job training, while undereducated workers are more likely to be admitted to such programs.